Hi, I'm having a problem that I thought had been solved, only for it to rear its head again recently. Some months ago, I began to experience a problem with my computer where, at any time, my screens would suddenly shut off and the fans in my computer would audibly ramp up. If I was in a discord call, I would still be able to talk to people for a short time before my computer would either turn off or I would manually shut it down. This happened at random. When I played games, or when I was just browsing the web. In August, I RMA'd my RTX 4080 back to MSI. They kept it for a week, then sent it back saying it had been repaired. I used the card for about a month and half since then, and the problem has not occurred since, at least until yesterday.
Back when this was a problem, Helldivers 2 was a real problem. It seemed to happen when playing this game more frequently than other instances. When trying to play yesterday, I had the game open for not even ten minutes when the same problem, happened. Then, it happened again less than two hours later, also while playing the game. After the first crash, I decided to play again with HWMonitor open to monitor my temps. Neither my GPU nor CPU ever got close to potentially dangerous or unstable territory.
I have tried some things already. As mentioned, I sent the 4080 back to MSI for repairs. This seemingly worked until playing a specific game, which leads me to think it isn't necessarily a hardware problem, but I'm not entirely sure. That being said, I still think that there is something about the 4080 that is causing this, as replacing the card with a 3070ti I had laying around resulted in me having no issues whatsoever. I have also contacted Cablemod, after reading that their StealthSense cables had a fix for the exact problem I was having. After receiving the new cable, the problem happened again within 30 minutes of playing Helldivers 2. I also want to note that this does NOT happen at all outside of this specific game, at least so far, and I play other modern, resource intensive games with no issues. This makes me think that the game itself is at least part of a problem, but my whole computer crashing in such a weird way still makes me think that it's a hardware problem.
I have also read online that it could be an issue with the seating of my power cable. I verified that the 3x8 cable powering the 4080 was properly seated on both ends. I removed all GPU drivers and then updated them to the latest version. Bios is up to date. I ran a Furmark stress test and a CPU-Z stress test simultaneously for fifteen minutes straight. While my CPU did eventually thermal throttle (12900KF moment), my 4080 never even cracked 80 degrees. My system showed no instability during the test. I have also tried a number of other stress tests for upwards of 30 minutes at a time with 100% CPU and GPU utilization, and my computer worked perfectly fine. I have also tried lowering the power limit on my GPU, underclocking my GPU, turning XMP on, and running the game with Dx11, none of which result in any changes.
Today, I used HWiNFO to try and collect some extra information, as due to the nature of the crash, no reports are generated. There are a couple things in here that stood out to me, including a hotspot temp of 110 degrees. However, during that reading the overall GPU was only at 80 degrees, and search results say that 125 degrees is the max hot spot temp for my card. There are also results from thermal, voltage reliability, and power limiters being at various points.
At this point, I'm considering contacting Corsair as perhaps there is something wrong with my power supply.
Full system specs:
i9-12900KF
32 GB DDR5 RAM
MSI RTX 4080 Ventus 3x OC
Windows 10 installed to an M.2 SSD
Corsair RM850x PSU
Back when this was a problem, Helldivers 2 was a real problem. It seemed to happen when playing this game more frequently than other instances. When trying to play yesterday, I had the game open for not even ten minutes when the same problem, happened. Then, it happened again less than two hours later, also while playing the game. After the first crash, I decided to play again with HWMonitor open to monitor my temps. Neither my GPU nor CPU ever got close to potentially dangerous or unstable territory.
I have tried some things already. As mentioned, I sent the 4080 back to MSI for repairs. This seemingly worked until playing a specific game, which leads me to think it isn't necessarily a hardware problem, but I'm not entirely sure. That being said, I still think that there is something about the 4080 that is causing this, as replacing the card with a 3070ti I had laying around resulted in me having no issues whatsoever. I have also contacted Cablemod, after reading that their StealthSense cables had a fix for the exact problem I was having. After receiving the new cable, the problem happened again within 30 minutes of playing Helldivers 2. I also want to note that this does NOT happen at all outside of this specific game, at least so far, and I play other modern, resource intensive games with no issues. This makes me think that the game itself is at least part of a problem, but my whole computer crashing in such a weird way still makes me think that it's a hardware problem.
I have also read online that it could be an issue with the seating of my power cable. I verified that the 3x8 cable powering the 4080 was properly seated on both ends. I removed all GPU drivers and then updated them to the latest version. Bios is up to date. I ran a Furmark stress test and a CPU-Z stress test simultaneously for fifteen minutes straight. While my CPU did eventually thermal throttle (12900KF moment), my 4080 never even cracked 80 degrees. My system showed no instability during the test. I have also tried a number of other stress tests for upwards of 30 minutes at a time with 100% CPU and GPU utilization, and my computer worked perfectly fine. I have also tried lowering the power limit on my GPU, underclocking my GPU, turning XMP on, and running the game with Dx11, none of which result in any changes.
Today, I used HWiNFO to try and collect some extra information, as due to the nature of the crash, no reports are generated. There are a couple things in here that stood out to me, including a hotspot temp of 110 degrees. However, during that reading the overall GPU was only at 80 degrees, and search results say that 125 degrees is the max hot spot temp for my card. There are also results from thermal, voltage reliability, and power limiters being at various points.
At this point, I'm considering contacting Corsair as perhaps there is something wrong with my power supply.
Full system specs:
i9-12900KF
32 GB DDR5 RAM
MSI RTX 4080 Ventus 3x OC
Windows 10 installed to an M.2 SSD
Corsair RM850x PSU